Research Article
Diagnosis of Chronic Kidney Disease Using Effective Classification Algorithms and Recursive Feature Elimination Techniques
Table 8
Comparison of the performance of our proposed system with previous studies.
| Previous studies | Accuracy % | Precision % | Recall % | F1-score % |
| Hore et al. [29] | 92.54 | 85.71 | 96 | 90.56 | Vasquez-Morales et al. [11] | 92 | 93 | 90 | 91 | Rady and Anwar [13] | 95.84 | 84.06 | 93.55 | 88.55 | Elhoseny et al. [19] | 85 | | 88 | 88 | Ogunleye and Wang [30] | 96.8 | | 87 | 93 | Khan et al. [31] | 95.75 | 96.2 | 95.8 | 95.8 | Chittora et al. [32] | 90.73 | 83.34 | 93 | 88.05 | Jongbo et al. [33] | 89.2 | 97.72 | 97.8 | | Harimoorthy and Thangavelu [34] | 66.3 | 65.9 | 65.9 | | Proposed model (random forest) | 100 | 100 | 100 | 100 | Proposed model (decision tree) | 99.34 | 98.68 | 100 | 99.17 | Proposed model (KNN) | 98.33 | 100 | 97.37 | 98.67 | Proposed model (SVM) | 97.3 | 94.74 | 92 | 96.67 |
|
|